Many everyday rhythmic activities such as breathing, chewing, and locomoting, are programed in part by neuronal networks called central pattern generators (CPGs). Mechanistic analysis of CPGs has been especially fruitful in the CPGs of invertebrates, which have relatively few neurons and can therefore be precisely defined in terms of identified neurons and specific synaptic connections. CPG networks of invertebrates have become proxies for experimental and theoretical analyses of how brain networks reliably yet flexibly process sensory information or program motor output. Our detailed analyses of the leech heartbeat CPG continue to contribute substantially to the organizing principles by which we now understand network function. Theoretical studies and experimental analysis in different networks and species have shown that the intrinsic membrane properties of the neurons and the strengths of their synaptic connections show 2-5 fold animal-to-animal variability, yet networks still produce functional output. Nevertheless, there are clear examples - including from our own work - showing that individual variation in synaptic and intrinsic properties do have functional consequences that manifest in network performance or susceptibility to perturbation. These studies imply that to understand fundamentally a neuronal network, we must strive to gather complete data from individuals because networks from individuals, while functionally stereotyped, may have different mechanistic underpinnings at the level of membrane currents and synaptic connections with discernible functional consequences. In light of this variation, an ensemble modeling approach is necessary to generate and test hypotheses about living networks. In this approach, multiple model instances (with different parameters) generated by evolutionary algorithms or brute force parameter variation are used. A prerequisite for this approach is experimental studies that determine parameter variation in the living system across a variety of networks. Our experiments will integrate computational and neurophysiological approaches to elucidate basic mechanisms of network function and the impact of individual variation in network parameters. Moreover, we will use individual variation as an innovative tool to probe network function. The leech heartbeat CPG is analogous to spinal CPGs that produce coordinated rhythmic activity in motor neurons, and its relative simplicity and superb accessibility allow a level of cellular analysis not currently possible in spinal cord. We will analyze the impact of inter-individual variation in CPG input and motor neuron intrinsic properties on motor output and how movements reflect this variation. These studies will provide basic mechanistic insights into network function in individuals, which will be useful in the study of spinal and brain networks and can lead to therapies for spinal cord and brain injury and disorders affecting motor performance.